General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

Users may download and print one copy of any publication from the public portal for the purpose of private study or research.

You may not further distribute the material or use it for any profit-making activity or commercial gain

You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from orbit.dtu.dk on: Aug 01, 2020

Deposit Shedding in Biomass-Fired Boilers: Shear Adhesion Strength Measurements

Laxminarayan, Yashasvi; Jensen, Peter Arendt; Wu, Hao; Jappe Frandsen, Flemming; Sander, Bo;Glarborg, Peter

Published in:Energy and Fuels

Link to article, DOI:10.1021/acs.energyfuels.7b01312

Publication date:2017

Document VersionPeer reviewed version

Link back to DTU Orbit

Citation (APA):Laxminarayan, Y., Jensen, P. A., Wu, H., Jappe Frandsen, F., Sander, B., & Glarborg, P. (2017). DepositShedding in Biomass-Fired Boilers: Shear Adhesion Strength Measurements. Energy and Fuels, 31(8), 8733-8741. https://doi.org/10.1021/acs.energyfuels.7b01312

1

Deposit Shedding in Biomass-fired Boilers: Shear

Adhesion Strength Measurements

Yashasvi Laxminarayan* †, Peter Arendt Jensen †, Hao Wu †, Flemming Jappe Frandsen †, Bo

Sander §, Peter Glarborg †

† Department of Chemical and Biochemical Engineering, Technical University of Denmark,

Søltofts Plads 229, 2800 Kgs. Lyngby, Denmark

§ DONG Energy A/S, Kraftsværksvej 53, Skærbæk, DK-7000, Fredericia, Denmark

Keywords: biomass, ash, boiler, adhesion strength, shedding, sintering

*Corresponding author e-mail id: ylax@kt.dtu.dk

Abstract: Ash deposition on boiler surfaces is a major problem encountered in biomass

combustion. Timely removal of ash deposits is essential for optimal boiler operation. In order to

improve the understanding of deposit shedding in boilers, this study investigates the adhesion

strength of biomass ash from full-scale boilers, as well as model fly ash deposits containing KCl,

K2SO4, CaO, CaSO4, SiO2, K2CO3, Fe2O3, K2Si4O9 and KOH. Artificial biomass ash deposits were

prepared on superheater tubes, and sintered in an oven with temperatures ranging from 500°C to

1000°C. Subsequently, the deposits were sheared off by an electrically controlled arm, and the

corresponding adhesion strength was measured. The effect of sintering temperature, sintering time,

deposit composition, thermal shocks on the deposit and steel type was investigated. The results

2

reveal that adhesion strength of ash deposits is dependent on two factors: ash melt fraction, and

corrosion occurring at the deposit-tube interface. Adhesion strength increases with increasing

sintering temperature, sharply increasing at the ash deformation temperature. However, sintering

time, as well as the type of steel used, does not have a significant effect under the investigated

conditions. Addition of compounds which increase the melt fraction of the ash deposit, typically

by forming a eutectic system, increases the adhesion strength, whereas addition of inert compounds

with a high melting point decreases the adhesion strength. Furthermore, the study indicated that

sulphation of ash deposits leads to an increase in adhesion strength, while cooling down the

deposits after sintering decreases the adhesion strength. Finally, it was observed that adhesion

strength data follows a log-normal distribution.

Introduction

One of the major operational problems encountered in biomass-fired boilers is the formation of

ash deposits on boiler surfaces. Ash deposition hinders the efficiency of heat transfer to the steam

cycle1 and may completely block flue gas channels in severe cases, causing expensive unscheduled

boiler shutdowns. Furthermore, ash deposits may cause severe corrosion of boiler surfaces.2

Therefore, timely and effective removal of ash deposits is essential for optimal boiler operation.

Natural as well as artificially induced shedding of ash deposits may be caused by several

mechanisms including erosion, debonding, molten slag flow, and thermal and mechanical stresses

in the deposits.3 Full-scale investigations have revealed that debonding is the dominant mechanism

for shedding of dense and hard deposits in biomass boilers,4 occurring when the generated stress

(e.g. by soot-blowing or due to the inherent weight of the deposit) exceeds the adhesion strength

at the deposit-tube interface.1 Hence, quantification of the adhesion strength of ash deposits is

crucial for understanding deposit shedding, and for optimizing artificial removal of deposits (e.g.

3

by soot-blowing or application of thermal shocks). Sootblowing in boilers produces both lateral

(lift) and longitudinal (drag) forces on deposits,5 highlighting the importance of understanding the

shear as well as tensile adhesion strength of ash deposits. Additionally, the adhesion strength at

the interface is dependent on the contact area between the steel tube and the innermost layer of the

ash deposit.5 The innermost layer of biomass ash deposits is primarily formed by heterogeneous

condensation, or homogeneous/heterogeneous nucleation and subsequent thermophoretic

deposition of alkali salts,6,7 with their composition typically dominated by KCl and K2SO4.8

Previous studies have investigated the adhesion strength of deposits for coal ash9-11 as well as

ash from kraft recovery boilers.5 Other studies have tried to quantify the inherent compression and

bend strength of sintered ash deposits.12-16 However, there is a lack of understanding of the

adhesion strength of biomass ash deposits to boiler surfaces. The literature lacks a detailed

investigation, describing the effect of various parameters, such as sintering temperature, chemical

composition and sintering time, on the adhesion strength of biomass ash deposits.

The present work quantifies the shear adhesion strength of biomass ash and salt rich deposits in

a laboratory oven, in order to determine the effect of gas and steel surface temperature, deposit

chemical composition, sintering duration, steel type and thermal shocks brought about by a rapid

change in sintering temperature. The study simulates the conditions present at the deposit-tube

interface, under different deposit properties and boiler conditions. Apart from providing a better

fundamental understanding of deposit shedding, the outcome of this study may facilitate boiler

operation by recommending boiler conditions for minimizing the formation of strong deposits.

Furthermore, the data obtained from this study may be used to optimize soot-blowing in boilers.

Experimental Section

4

Materials

Experimental analysis was carried out using fly ash obtained from the electrostatic precipitator

/ bag filter of a straw-fired grate boiler (Avedøreværket unit 2, 100 MWth), a wood-fired suspension

boiler (Avedøreværket unit 2, 800 MWth), and a straw + wood co-fired suspension boiler

(Amagerværket unit 1, 350 MWth). The fly ash properties are provided in Table 1. While the straw

fly ash is rich in K and Cl, the wood fly ash and the straw + wood co-fired fly ash are rich in Ca

and Si. As a result, the ash deformation temperature,17 which is the temperature at which the ash

first softens and therefore becomes sticky,18 of straw fly ash is low (640°C), whereas the ash

deformation temperature of wood fly ash and straw + wood co-fired fly ash is rather high (1240°C

and 1220°C). Additionally, model fly ash deposits were prepared using mixtures of KCl (Sigma

Aldrich, CAS number: 7447-40-7), K2SO4 (Sigma Aldrich, CAS number: 7778-80-5), K2CO3

(Sigma Aldrich, CAS number: 584-08-7), CaO (Sigma Aldrich, CAS number: 1305-78-8), CaSO4

(Alfa Aesar, CAS number: 7778-18-9), SiO2 (Sigma Aldrich, CAS number: 60676-86-0), Fe2O3

(Sigma Aldrich, CAS number: 1309-37-1), K2Si4O9 (Alfa Aesar, CAS number: 1312-76-1) and

KOH (Sigma Aldrich, CAS number: 1310-58-3), in order to understand the effect of different

components constituting a typical biomass fly ash. The melting point / eutectic point / glass

transition temperature of the model fly ash compounds is provided in Table 2. Each of the different

components was milled and sieved individually to obtain a particle size distribution bounded by

32 μm and 90 μm. However, it should be noted that fly ash in boilers typically form a bimodal

particle size distribution, consisting of sub-micron particles, as well as larger particles (~10 µm -

200 µm)1,6. Although the deposits prepared in this study do not contain any sub-micron particles,

it is ensured that the particle size lies within the second peak of the characteristic bimodal size

5

distribution. Since KCl and K2SO4 are the major species found in the inner layer of typical biomass

ash deposits,8 all investigated model fly ash deposits contained KCl and K2SO4.

Experiments were carried out using 3 different types of steel, TP347HFG (Salzgitter

Mannesmann), 316SS (Sandvik) and 3R69BT (Sandvik), as well as tubes made from pure iron.

The tubes had an outer diameter of 38 mm, and a thickness of 5 mm. The chemical composition

of the steel tubes is provided in Table 3. The addition of Cr, Mo and Mn in steel reduces oxide

scale growth,19,20 improving overall corrosion resistance,21 while Ni acts as a deterrent for Cl

induced corrosion.22 Pre-oxidation of steel tubes is beneficial for hindering corrosion,23,24 and

provides superior replication of operational boiler tubes.25,26 Thermogravimetric analysis of the

steel tubes at 600°C revealed that majority of the oxidation occurs in first few hours, after which

the rate of oxidation significantly slows down (see Figure 1). Therefore, the tubes were pre-

oxidized for 24 hours at 600°C prior to conducting experiments.

Sample preparation

In order to obtain tightly packed and adherent deposits, the ash particles were mixed with a 50%

isopropanol solution to prepare a thick slurry, and molded into a cubical shaped deposit on the

surface of the tube, using a Teflon mold (see Figure 2). The deposits were 15 mm x 15 mm x 10

mm, (WxDxH) in size, leading to a contact surface area of 223 mm2. The use of deposit slurries is

in accordance with EU guidelines2,27,28 for high temperature corrosion testing, providing a better

representation of deposits in power plants. However, it should be noted that the deposit formation

process and the typical particle size distribution of fly ash in boilers is different from the samples

prepared in this study.1,6,8

Deposit sintering and adhesion strength measurement

6

The deposits were heated up and sintered inside an oven for a fixed duration. A purge air flow

of 15 NL/min was injected into the oven, to protect the oven heating elements from corrosion.

After sintering, the deposits were cooled down to the required measurement temperature at a

rate of 15°C/min, subsequently followed by shear adhesion strength measurements. An electrically

controlled arm was used to de-bond the artificial ash deposit from the superheater tube, as shown

in Figure 2. The arm was controlled using a linear actuator, and the corresponding force applied

on the ash deposit was measured using a load cell. Shear adhesion strength was calculated by

dividing the measured force by the contact area between the deposit and the superheater tube.

Standard experiments were performed by sintering the deposits at 650°C for 4 hours, while the

adhesion strength was measured at 600°C. These parameters were chosen providing consideration

to typical sintering temperatures of the inner layers of the deposit,8 typical boiler steam

temperatures,29-32 temperature gradients across the steel tube, resulting in the steel surface

temperature to be 20°C - 50°C higher than the steam temperature,8,33,34 reasonable experimental

time and the deposit formation process.6,31 In order to account for the scatter observed while

measuring adhesion strength, measurements were conducted on at least 4 deposit samples for each

instance of experimental conditions.

Selected samples were analyzed using Scanning Electron Microscopy to observe the deposit-

tube interface. The steel tubes, along with deposits, were cast in epoxy and polished, without any

exposure to water, thereby preventing any dissolution, recrystallization and removal of salts.

Results and Discussion

Effect of sintering temperature

7

Figure 3 shows the effect of sintering temperature on adhesion strength. Experiments were

performed with pure KCl, as well as three different boiler fly ashes (see Table 1). It can be observed

that adhesion strength increases with increasing temperature, with a sharp increase near the melting

point / ash deformation temperature,17 i.e., 640°C for straw fly ash, and 770°C for KCl.

A sharp increase in adhesion strengths for wood fly ash and the straw + wood co-fired fly ash

has not been observed in this study, due to their high ash deformation temperatures, 1220°C and

1240°C respectively (see Figure 3). Furthermore, it should be noted that increasing the temperature

of the oven to temperatures significantly higher than the melting point (or ash deformation

temperature) led to completely molten deposits, whose adhesion strength could not be measured.

Effect of eutecticity

The constituents of fly ash typically form eutectic systems,3,35 leading to melt formation at

temperatures lower than the melting point of the individual components. The aforementioned

results seem to indicate that the adhesion strength of an ash deposit is dependent on its melting

point. Previous studies in literature have indicated that the melt fraction of the deposit, especially

at the deposit-tube interface, influences its adhesion strength.4,5 In order to better understand this

phenomenon, experiments were performed with model fly ash compounds containing KCl and

K2SO4. KCl and K2SO4, with individual melting points of 770°C and 1069°C, form a eutectic

system with a eutectic temperature of 690°C. The eutectic temperature was calculated using the

software, FactSage.36,37 However, other experimental studies have identified melt formation at

683°C for a 50 wt.% KCl-K2SO4 system.38 In the experiments, the amount of K2SO4 in KCl was

varied at 650°C, and the corresponding results are shown in Figure 4, along with the KCl-K2SO4

phase diagram. The phase diagram was obtained using FactSage.

8

The results indicate that while pure substances do not have much adhesion strength at 650°C,

mixing of the components causes a large increase in the adhesion strength. Since KCl and K2SO4

form a eutectic system, mixing of the two components leads to an increase in the melt fraction of

the deposit. Therefore, it can be inferred that a higher melt fraction at the deposit-tube interface

leads to a higher adhesion strength. However, the experiments were carried out at 650°C, which is

lower than the eutectic temperature of the KCl-K2SO4 system. This indicates the presence of a

secondary phenomenon influencing deposit adhesion strength, which has been explored in the

following section by conducting a SEM analysis of the deposit-tube interface.

SEM analysis of the deposit-tube interface

In order to determine the morphology of the deposits at the deposit-tube interface, SEM analysis

of the interface was carried out for the model fly ash deposit containing KCl and K2SO4 (50 wt.%).

The analysis revealed the formation of a dense, partially molten layer at the interface, as seen in

Figure 5. As the temperature increases, corrosion starts to occur at the interface. As a result,

corrosion products, such as Fe/Cr chlorides, oxides, chromates, etc., are formed.2,39-43 Most of the

corrosion products form a complex eutectic system with the components present in the deposit.44,45

This leads to a lower eutectic temperature at the interface, compared to the outer layers of the

deposit. The partially molten layer causes increased surface wetting and adsorption,46 leading to

high surface adhesion.

It should be noted that debonding always occurred in the corrosion layer throughout all

experiments, exposing a fresh layer of steel tube after deposit removal.

Effect of composition

9

Experiments were conducted with model fly ash deposits to understand the role of different

components present in a typical biomass fly ash. The model fly ash deposits were made up of

particles larger than 32 µm and smaller than 90 µm.

The results highlight the effect of sulphation on adhesion strength, as seen in Figure 6. The

deposit containing KCl and K2SO4 (50 wt.%) exhibited much higher adhesion strength compared

to a deposit containing pure KCl. Similarly, the deposit containing KCl, K2SO4 and CaSO4 (33

wt.% each) showed a higher adhesion strength than the deposit containing KCl, K2SO4 and CaO.

The increase in adhesion strength of deposits containing sulphur can be attributed to the fact that

KCl-K2SO4 and KCl-K2SO4-CaSO4 form a eutectic system (see Table 2). Therefore, sulphation

lowers the eutectic/deformation temperature of the ash deposit, increasing melt fraction, and

thereby increasing adhesion strength.

In boilers, KCl can undergo sulphation in the gas phase prior to deposition,47 or in solid phase

after deposition on boiler surfaces.48 While gas phase sulphation is faster than solid phase

sulphation, the deposit is exposed to the flue gas for a longer period of time,48 making both

sulphation mechanisms relevant. Similarly, CaO can undergo sulphation to form CaSO4.49,50

Sulphation of KCl in deposits can occur as a gas-solid or gas-liquid reaction by SO251 or SO3,

52

as shown in the following equations.

2 KCl + SO2 + ½ O2 + H2O → K2SO4 + 2 HCl

2 KCl + SO3 + H2O → K2SO4 + 2 HCl

Iron oxide may catalytically convert SO2 to SO3,45,53 or react with SO2 to form Fe(III) sulphites

or sulphates,54,55 thereby catalyzing the overall sulphation reaction and increasing the

concentration of K2SO4 near the steel surface.

The present results provide evidence that sulphation may result in an increase in adhesion

strength at the investigated conditions. However, these results are not conclusive, since sulphate-

10

forming reactions occurring inside the deposit have not been explored in the conducted

experiments. Further investigation of deposit sulphation is required to completely understand the

influence of the overall sulphation process on deposit adhesion strength. Nevertheless, it is

speculated that reactions occurring between the deposit and the flue gas may contribute to adhesion

strength variations in boilers.

Furthermore, the results portray the effect of CaO, SiO2, K2CO3, Fe2O3, K2Si4O9 and KOH (see

Figure 6). While Ca and Si are widely present in biomass ash deposits, the presence of K2CO3 has

been identified in only a few studies in literature.56,57 The addition of CaO to a model fly ash

deposit containing KCl-K2SO4 decreased its adhesion strength. CaO does not form a eutectic melt

with the KCl-K2SO4 system, effectively reducing the melt fraction, and thereby decreasing the

adhesion strength. However, the addition of SiO2 does not seem to significantly affect the adhesion

strength under the conditions examined.

The addition of K2CO3 to the model fly ash deposit containing KCl and K2SO4 considerably

increased the adhesion strength. Addition of K2CO3 decreases the eutectic temperature of the KCl-

K2SO4 system (see Table 2), increasing the melt fraction of the ash deposit at 650°C, and thereby

increasing the adhesion strength. Furthermore, K2CO3 may react with the steel, leading to the

formation of a potassium-chromium compound, most likely K2CrO4,41,58,59 which forms a low-

temperature melt with KCl,60 further increasing the melt fraction at the deposit-tube interface.

Moreover, the addition of Fe2O3 significantly increased the adhesion strength of the ash deposits,

bolstering the aforementioned theory correlating corrosion with high adhesion strength. Apart

from decreasing the melting point of the mixture (see Table 2), presence of Fe2O3 in the deposit

may cause increased formation of corrosion intermediates, such as FeCl2 or FeCl3, according to

the following proposed reaction. The reaction mechanism has been verified using Factsage.37

11

Fe2O3 + 6 KCl ⇌ 2 FeCl3 + 3 K2O

Moreover, in full-scale boilers, where HCl present in the flue gas may be oxidized to Cl2, the

following reaction may occur, leading to the formation of FeCl2.22,61,62

Fe2O3 + 2 Cl2 ⇌ 2 FeCl2 + 1.5 O2

Since FeCl2, as well as FeCl3, forms a eutectic system with the ash deposit, the corresponding

increase in melt fraction results in an increase in adhesion strength.

A similar increase in adhesion strength is observed when K2Si4O9 is added to the KCl-K2SO4

system. The presence of alkali silicates has been identified in mature and sintered deposits in straw-

fired boilers.8 K2Si4O9 is known to form a glass phase at high temperatures, gradually decreasing

in viscosity with increasing temperature.63 Analysis of the K2Si4O9 samples using Differential

Scanning Calorimetry revealed that K2Si4O9 has a glass transition temperature of 650°C (see

Figure 7). The formation of a semi-molten glass phase causes an increase in surface wetting and

increased adhesion of the deposit to the steel tube.

The presence of KOH in deposits has been postulated in a few studies in literature.56,62 In the

present study, it was observed that even the addition of a small amount of KOH (2.5 wt.%) to the

model fly ash deposit causes a large increase in adhesion strength. This can directly be attributed

to the low melting point of KOH (360°C) and the formation of a eutectic system with KCl-K2SO4

(see Table 2), causing increased melt formation and adhesion strength.

From this section, it can be concluded that addition of compounds which increase the melt

fraction of the ash deposit, usually by forming a eutectic system, increases the adhesion strength.

However, addition of inert compounds with a high melting point, such as CaO (melting point of

2572°C), decreases the adhesion strength.

Effect of sintering time

12

Sintering time seems to have a negligible effect on adhesion strength up to 24 hours at the

investigated conditions, as seen in Figure 8. It should be noted that the all experiments are

subjected to an additional 30 minutes of heating time prior to sintering, and 5 minutes for strength

measurement after sintering.

The results suggest that the initial, partially molten corrosion layer is formed rather quickly, and

significant changes in adhesion strength do not occur after the formation of the initial corrosion

layer at the interface within 24 hours. Several studies in literature indicate that the onset of

corrosion is typically within a few minutes, and the rate of corrosion decreases exponentially over

time.2,39,64 The marginal change in melt fraction due to increasing corrosion is not significant

enough to observe reliable changes in adhesion strength. However, further investigation is required

prior to arriving at conclusions, especially considering that sintering in boilers may occur for

longer durations.

Nevertheless, an increase in deposit adhesion strength may be observed in boilers due to

sintering caused by reactions occurring in the deposit, e.g., sulphation,65 which have not been

investigated in this study. Sulphation does not occur in the experimental setup, due to the absence

of SO2 in the gas stream.

Effect of thermal shocks

Application of thermal shocks to induce deposit shedding is a technique commonly used to

remove heavily sintered deposits from superheater tubes.13,66 This study further investigates the

effect of thermal shocks by cooling down the deposit after sintering. Deposits were cooled down

at a rate of 15°C/min.

As seen in Figure 9, cooling down the deposits results in a decrease in adhesion strength.

Thermal stresses are induced at the deposit-tube interface, owing to differences in the thermal

13

expansion coefficients between the deposit/corrosion layer and the steel tube.66 As a result, cracks

may develop at the interface, leading to a decrease in adhesion strength.

Effect of steel type

In order to understand the effect of the type of steel used, experiments were carried out using a

model fly ash deposit containing KCl-K2SO4 (50 wt.%) on 3 different types of steel as well as pure

iron tubes. Experiments were carried out for 4 hours at 650°C.

The results indicate that the type of steel used does not have a strong influence on the adhesion

strength at the investigated conditions, considering the scatter in data (see Figure 10). Previous

studies have shown that KCl induces corrosion at the steel surface, irrespective of the type of

steel,58 although the depth of the corrosion layer might be different. The results seem to indicate

that while the presence of corrosion causes high adhesion strength, the depth of the corrosion layer

is not a major factor influencing adhesion strength, especially considering that the onset of the

corrosion layer is typically within a few minutes.2,39,64

However, the adhesion strength of deposits to pure iron tubes appears to be slightly higher, when

compared to the investigated steels, indicating that the presence of corrosion inhibiting elements

in steel might play a role in influencing adhesion strength. Further investigation, spanning over a

larger range of steel types, is required prior to arriving at conclusions.

Analysis of scatter in adhesion strength data

In order to better understand the significant scatter observed in the data, 24 experiments were

conducted using KCl-K2SO4 (50 wt.%) deposits. The experiments reveal that adhesion strength

data roughly follows a log-normal distribution, as seen in Figure 11. This is similar to observations

made using deposits from kraft recovery boilers.5 Moreover, experiments conducted in full-scale

biomass-fired boilers indicate similar trends.4

14

The stochastic nature of debonding has significant implications on deposit shedding in boilers.

The results suggest that even though soot-blowing may remove the majority of the deposits, the

strongly adherent deposits might not be removed. Subsequent accumulation of strong deposits

probably results in the eventual fouling of boiler surfaces.5

The adhesion strength of biomass ash deposits observed in this study is comparable in magnitude

to coal ash deposits from lab-scale investigations,9-11 as well as biomass ash deposits from full-

scale studies,4 as shown in Table 4. However, previous lab-scale investigations indicate that

deposits from kraft recovery boilers are more strongly adherent, when compared to biomass and

coal ash deposits.5

Practical application of the study

The results allow better understanding of the process of deposit shedding, both qualitatively and

quantitatively. Furthermore, the obtained data may be used to develop a tool for analyzing the

effect of fuel composition on adhesion strength, and suggesting boiler operating conditions to

prevent the formation of strong deposits. For example, the study identifies that maintaining steel

temperatures below the ash deformation temperature results in the formation of weaker deposits.

Furthermore, the study quantifies the degree of thermal shocks needed to weaken the strongly

adherent deposits. Moreover, the study analyzes the effect of composition of the fly ash, which

could be used to estimate fuel quality. However, further work is required prior to arriving at

conclusions.

Additionally, the obtained data may be used to optimize soot-blowing in boilers by

recommending soot-blowing frequencies and pressures based on the fuel and operating conditions.

This may be done by modelling the log-normal distribution of adhesion strength data,

15

incorporating the effect of deposit composition, flue gas temperature and steam temperature.

However, further experimental work is required for the development of a detailed model.

Conclusions

This study investigated the shear adhesion strength of biomass ash deposits from full-scale

boilers, as well as model fly ash deposits containing KCl, K2SO4, CaO, CaSO4, SiO2, K2CO3,

Fe2O3, K2Si4O9 and KOH. Deposits were prepared on superheater tubes, and sintered in a

laboratory oven. The effect of sintering temperature, sintering time, deposit composition, thermal

shocks on the deposit and steel type was investigated.

Increasing sintering temperatures resulted in higher adhesion strengths, with a sharp increase

observed near the ash deformation temperature / melting point. Sintering time did not significantly

affect adhesion strengths up to 24 hours at 650°C, using a model fly ash deposit containing KCl-

K2SO4 (50 wt.%). Furthermore, it was substantiated that cooling down the deposit after sintering

reduces the adhesion strength, due to thermal stresses induced at the deposit-tube interface.

Deposits containing sulphates showed increased adhesion strengths, indicating that sulphation

may cause the formation of stronger deposits. The addition of K2CO3, Fe2O3, K2Si4O9 and KOH

to the model fly ash deposit increased the ash melt fraction at the deposit-tube interface, thereby

increasing the adhesion strength, whereas the addition of CaO decreased the ash melt fraction,

thereby decreasing the adhesion strength.

Furthermore, the type of steel used did not seem to have a considerable effect on the adhesion

strength. Finally, experiments revealed that adhesion strength data roughly follows a log-normal

distribution.

16

This study identified that adhesion strength of ash deposits is dependent on two factors: ash melt

fraction, and corrosion occurring at the deposit-tube interface. A higher ash melt fraction at the

deposit-tube interface leads to an increase in adhesion strength. Corrosion occurring at the

interface leads to the formation of corrosion products, which form a eutectic system with the inner

layer of deposit and increase the local melt fraction, thereby increasing the adhesion strength.

17

FIGURES

Figure 1. Thermogravimetric analysis of steel used (TP347HFG) exposed to air at 600°C. Most

of the oxidation occurs within the first few hours, after which the oxidation rate significantly slows

down. Sample mass of 2059 mg, heating rate of 10 K/min.

Figure 2. Experimental setup for adhesion strength measurements. The superheater steel tube is

placed inside the oven while the load cell is outside the oven. The actuator arm shears off the

artificial ash deposit and the load cell measures the corresponding adhesion strength. Image not to

scale.

18

Figure 3. Effect of sintering temperature on shear adhesion strength for KCl and biomass fly ashes.

Shear adhesion strength increases sharply near the melting point / ash deformation temperature.17

Deposits sintered for 4 hours, measured at 600°C, TP347HFG steel pre-oxidized for 24 hours,

average of 4 data points.

19

Figure 4. Effect of varying concentration of K2SO4 in KCl on shear adhesion strength. Deposits

sintered at 650°C for 4 hours, measured at 600°C, TP347HFG steel pre-oxidized for 24 hours,

average of 4 data points (24 for 50 wt.%). Mixing of KCl and K2SO4 causes an increase in adhesion

strength.

20

Figure 5. SEM image of deposit-tube interface. KCl-K2SO4 (50 wt.%) deposit, sintered at 650°C

for 4 hours, TP347HFG steel pre-oxidized for 24 hours. Partially molten corrosion layer observed

at the deposit-tube interface.

21

Figure 6. Effect of composition on adhesion strength using model fly ash compounds. Deposits

sintered at 650°C for 4 hours, measured at 600°C, TP347HFG steel pre-oxidized for 24 hours, 4

data points (24 for KCl + K2SO4). All compositions in weight %.

22

Figure 7. Differential Scanning Calorimetry analysis of K2Si4O9. The silicate forms a glassy

phase, with a glass transition temperature of 650°C. Sample mass of 10.5 mg, heating rate of 10

K/min.

Figure 8. Effect of sintering time on adhesion strength. KCl-K2SO4 (50 wt.%) deposit, sintered at

650°C, measured at 600°C, TP347HFG steel pre-oxidized for 24 hours, 4 data points (24 for 4

hours).

23

Figure 9. Effect of strength measurement temperature on adhesion strength. KCl-K2SO4 (50 wt.%)

deposit, sintered at 650°C for 4 hours, TP347HFG steel pre-oxidized for 24 hours, 4 data points

(24 for 600°C)

Figure 10. Effect of steel type on adhesion strength. KCl-K2SO4 (50 wt.%) deposit, sintered at

650°C for 4 hours, measured at 600°C, steels pre-oxidized for 24 hours, 4 data points (24 for

TP347HFG)

24

Figure 11. Log-normal distribution of adhesion strength data. KCl-K2SO4 (50 wt.%) deposit,

sintered at 650°C for 4 hours, measured at 600°C, TP347HFG steel pre-oxidized for 24 hours, 24

data points

25

TABLES

Table 1. Composition, particle size and melting point analysis of the investigated fly ashes

Elemental composition (wt.

%, dry basis)

Straw fly ash,

grate fired

Straw + wood co-

fired fly ash,

suspension fired

Wood fly ash,

suspension fired

Al -- 2 2.13

Ca 1.3 20 20.8

Cl 19 1.3 0.2

Fe 0.044 1.4 1.73

K 43 9.1 6.26

Mg 0.12 3.3 3.22

Na 0.9 0.9 0.43

P -- 1.4 1.09

S 7.9 1.5 1.08

Si 1.1 12 17.7

Ti -- 0.14 --

Mn 0.059 -- --

Deformation temperature17 (°C) 640 1240 1220

Hemispherical temperature17

(°C) 640 1250 1230

Fluid temperature17 (°C) 760 1260 1240

Median particle size (μm) 51.7 44.5 34.7

26

Table 2. Eutectic temperature / melting point / glass transition temperature of the investigated

model fly ash compounds. Data obtained from multiple sources8,37,38,67

Composition Eutectic

temperaturea /

melting pointb /

glass transition

temperaturec (°C)

KClb 770

KCl + K2SO4a 690

KCl + K2SO4+ CaOa 690

KCl + K2SO4+ CaSO4a 644

KCl + K2SO4+ SiO2a 690

KCl + K2SO4+ K2CO3a 580

KCl + K2SO4+ Fe2O3a 577

KCl + K2SO4+ K2Si4O9c 650

KCl + K2SO4+ KOHa 288

Table 3. Composition of the investigated steel tubes

Steel type Cr

(wt.%)

Ni

(wt.%)

Fe

(wt.%)

Others (wt.%)

Iron

100

316SS 16-18 10-14 balance C=0.08, Si=0.75, Mn=2, P=0.045, S=0.03, Mo=2.5

TP347HFG 17-20 9-13 balance C=0.08, Si=0.75, Mn=2, P=0.04, S=0.03,

Nb+Ta=1

3R69BT 17.5 12.5 balance C=0.03, Si=0.4, Mn=1.7, P=0.03, S=0.015,

Mo=2.2

Table 4. Adhesion strength of different types of deposits.

27

Deposit type Adhesion

strength (kPa)

Biomass ash deposits,

current lab-scale

investigation

1 – 350

Biomass ash deposits, full-

scale investigations4

20 – 250

Coal ash deposits, lab-

scale investigations9-11

35 – 350

Deposits from kraft

recovery boilers, lab-scale

investigations5

1000-16000

Acknowledgements

This work is part of the project, ‘Flexible use of Biomass on PF fired power plants’, funded by

Energinet.dk through the ForskEL program, DONG Energy and DTU.

References

(1) Zbogar, A.; Frandsen, F.; Jensen, P. A.; Glarborg, P. Prog. Energy Combust. Sci. 2009, 35,

31-56.

(2) Okoro, S. C.; Montgomery, M.; Frandsen, F. J.; Pantleon, K. Energy Fuels 2014, 28, 6447-

6458.

(3) Frandsen, F. Ash formation, deposition and corrosion when utilizing straw for heat and

power production; Technical University of Denmark, Department of Chemical Engineering,

2010.

28

(4) Bashir, M. S.; Jensen, P. A.; Frandsen, F.; Wedel, S.; Dam-Johansen, K.; Wadenback, J.

Energy Fuels 2012, 26, 5241-5255.

(5) Kaliazine, A.; Cormack, D.; Ebrahimi-Sabet, A.; Tran, H. J. Pulp Paper Sci. 1999, 25, 418-

424.

(6) Zhou, H.; Jensen, P. A.; Frandsen, F. J. Fuel 2007, 86, 1519-1533.

(7) Baxter, L. L. Biomass Bioenergy 1993, 4, 85-102.

(8) Jensen, P. A.; Frandsen, F. J.; Hansen, J.; Dam-Johansen, K.; Henriksen, N.; Hörlyck, S.

Energy Fuels 2004, 18, 378-384.

(9) Moza, A. K.; Austin, L. G. Fuel 1981, 60, 1057-1064.

(10) Moza, A.; Shoji, K.; Austin, L. J. Inst. Energy 1980, 53, 17-19.

(11) Raask, E. VGB Kraftwerkstech 1973, 53, 248-254.

(12) Attig, R.; Barnhart, D. A laboratory method of evaluating factors affecting tube bank fouling

in coal-fired boilers, in: Johnson, H. R.; Littler, D. J. The Mechanism of Corrosion by Fuel

Impurities; Butterworth: London, 1963; pp 173.

(13) Conn, R.; Jones, M. Comparison of low-rank coal ash sintering characteristics with pilot

plant ash fouling tendencies, in: Barrett, R. E. Third Engineering Foundation Conference on

Slagging and Fouling Due to Impurities in Combustion Gases; Copper Mountain: Colorado,

1984; pp 207-228.

29

(14) Cumming, I.; Joyce, W.; Kyle, J. J. Inst. Energy 1985, 58, 169-175.

(15) Yilu, S. Evaluating fouling and slagging characteristics of different Chinese coal ashes, in

:Third engineering foundation conference on slagging and fouling due to impurities in

combustion gases, in: Barrett, R. E. Third Engineering Foundation Conference on Slagging and

Fouling Due to Impurities in Combustion Gases; Copper Mountain: Colorado, 1984; pp 557-

592.

(16) Piroozmand, F.; Tran, H.; Kaliazine, A.; Cormack, D. Strength of recovery boiler fireside

deposits at high temperatures; Tappi Proceedings, 1998; pp 169-179.

(17) European Committee for Standardization, Solid biofuels - Method for the determination of

ash melting behaviour - Part 1: Characteristic temperatures method, CEN/TS 15370-1:2006.

(18) Gupta, S.; Wall, T.; Creelman, R.; Gupta, R. Fuel Process. Technol. 1998, 56, 33-43.

(19) Hamzah, M.; Yeo, W.; Fry, A.; Inayat-Hussain, J.; Ramesh, S.; Purbolaksono, J. Eng.

Failure Anal. 2013, 35, 380-386.

(20) Schmidt, D.; Schütze, M. Mater. Sci. Forum 2011, 696, 330-335.

(21) Yukitoshi, T.; Yoshikawa, K.; Daikoku, T.; Masuyama, F. J. Mater. Energy Syst. 1982, 4,

99-109.

(22) Zahs, A.; Spiegel, M.; Grabke, H. J. Corros. Sci. 2000, 42, 1093-1122.

(23) Okoro, S. C.; Kvisgaard, M.; Montgomery, M.; Frandsen, F. J.; Pantleon, K. Surf. Eng.

2016, 33, 428-432.

30

(24) Israelsson, N.; Engkvist, J.; Hellström, K.; Halvarsson, M.; Svensson, J.; Johansson, L.

Oxid. Met. 2015, 83, 29-53.

(25) Sabau, A.; Wright, I.; Shingledecker, J. Mater. Corros. 2012, 63, 896-908.

(26) Rizzi, M.; Guerrini, E.; Trasatti, S. Metall. Ital. 2012, 2, 29-35.

(27) Coleman, K.; Simms, N.; Kilgallon, P.; Oakey, J. Mater. Sci. Forum 2008, 595, 377-386.

(28) Grabke, H. J.; Meadowcroft, D. A working party report on guidelines for methods of testing

and research in high temperature corrosion; European Federation of Corrosion: Belgium, 1995.

(29) Bashir, M. S.; Jensen, P. A.; Frandsen, F.; Wedel, S.; Dam-Johansen, K.; Wadenbäck, J.;

Pedersen, S. T. Fuel Process. Technol. 2012, 97, 93-106.

(30) Wu, H.; Shafique Bashir, M.; Jensen, P. A. Full-scale ash deposition measurements at

Avedøre Power Plant unit 2 during suspension-firing of wood with and without coal ash

addition; Technical University of Denmark, Department of Chemical Engineering, 2012.

(31) Hansen, S. B. Model for Deposition Build-up in Biomass Boilers; Technical University of

Denmark, 2015.

(32) Montgomery, M.; Jensen, S.; Borg, U.; Biede, O.; Vilhelmsen, T. Mater. Corros. 2011, 62,

593-605.

(33) Lindberg, D.; Niemi, J.; Engblom, M.; Yrjas, P.; Laurén, T.; Hupa, M. Fuel Process.

Technol. 2016, 141, 285-298.

31

(34) Montgomery, M.; Karlsson, A.; Larsen, O. H. Mater. Corros. 2002, 53, 121-131.

(35) van Lith, S. C. Release of inorganic elements during wood-firing on a grate; Technical

University of Denmark, 2005.

(36) Bale, C.; Bélisle, E.; Chartrand, P.; Degterov, S.; Eriksson, G.; Hack, K.; Jung, I.; Kang, Y.;

Melançon, J.; Pelton, A. Calphad 2009, 33, 295-311.

(37) Bale, C.; Chartrand, P.; Degterov, S.; Eriksson, G.; Hack, K.; Mahfoud, R. B.; Melançon, J.;

Pelton, A.; Petersen, S. Calphad 2002, 26, 189-228.

(38) Arvelakis, S.; Jensen, P. A.; Dam-Johansen, K. Energy Fuels 2004, 18, 1066-1076.

(39) Sui, J.; Lehmusto, J.; Bergelin, M.; Hupa, M. Oxid. Met. 2014, 82, 437-456.

(40) Shinata, Y.; Nishi, Y. Oxid. Met. 1986, 26, 201-212.

(41) Pettersson, J.; Asteman, H.; Svensson, J.; Johansson, L. Oxid. Met. 2005, 64, 23-41.

(42) Spiegel, M. Mater. Corros. 1999, 50, 373-393.

(43) Lee, S. Y.; McNallan, M. J. Corrosion 1991, 47, 868-874.

(44) Viklund, P.; Hjörnhede, A.; Henderson, P.; Stålenheim, A.; Pettersson, R. Fuel Process.

Technol. 2013, 105, 106-112.

(45) Nielsen, H. P.; Frandsen, F. J.; Dam-Johansen, K. Energy Fuels 1999, 13, 1114-1121.

32

(46) Packham, D. E. Theories of fundamental adhesion, In: da Silva, L. F. M.; Öchsner, A.;

Adams, R. D. Handbook of Adhesion Technology; Springer: Berlin, 2011; pp 9-38.

(47) Iisa, K.; Lu, Y.; Salmenoja, K. Energy Fuels 1999, 13, 1184-1190.

(48) Sengeløv, L. W.; Hansen, T. B.; Bartolome, C.; Wu, H.; Pedersen, K. H.; Frandsen, F. J.;

Jensen, A. D.; Glarborg, P. Energy Fuels 2013, 27, 3283-3289.

(49) Borgwardt, R. H.; Bruce, K. R. AIChE J. 1986, 32, 239-246.

(50) Marsh, D.; Ulrichson, D. Chem. Eng. Sci. 1985, 40, 423-433.

(51) Henriksson, M.; Warnqvist, B. Ind. Eng. Chem. Process Des. Dev. 1979, 18, 249-254.

(52) Fielder, W.; Stearns, C.; Kohl, F. J. Electrochem. Soc. 1984, 131, 2414-2417.

(53) Jørgensen, T. L.; Livbjerg, H.; Glarborg, P. Chem. Eng. Sci. 2007, 62, 4496-4499.

(54) Fu, H.; Wang, X.; Wu, H.; Yin, Y.; Chen, J. J. Phys. Chem. C 2007, 111, 6077-6085.

(55) Baltrusaitis, J.; Cwiertny, D. M.; Grassian, V. H. Phys. Chem. Chem. Phys. 2007, 9, 5542-

5554.

(56) Wu, H.; Bashir, M. S.; Jensen, P. A.; Sander, B.; Glarborg, P. Fuel 2013, 113, 632-643.

(57) Blomberg, T. Mater. Corros. 2006, 57, 170-175.

(58) Kiamehr, S.; Dahl, K. V.; Montgomery, M.; Somers, M. A. Mater. Corros. 2015, 66, 1414-

1429.

33

(59) Lehmusto, J.; Skrifvars, B.; Yrjas, P.; Hupa, M. Fuel Process. Technol. 2013, 105, 98-105.

(60) Ishitsuka, T.; Nose, K. Corros. Sci. 2002, 44, 247-263.

(61) Okoro, S. C.; Kiamehr, S.; Montgomery, M.; Frandsen, F. J.; Pantleon, K. Mater. Corros.

2017, 68, 499-514.

(62) Folkeson, N.; Jonsson, T.; Halvarsson, M.; Johansson, L.; Svensson, J. Mater. Corros.

2011, 62, 606-615.

(63) Dickinson, J. E.; Scarfe, C. M.; McMillan, P. J. Geophys. Res. 1990, 95, 15675-15681.

(64) Karlsson, S.; Jonsson, T.; Hall, J.; Svensson, J.; Liske, J. Energy Fuels 2014, 28, 3102-3109.

(65) Hurley, J. P.; Benson, S. A.; Energy Fuels 1995, 9, 19-30.

(66) Raask, E. Mineral impurities in coal combustion: behavior, problems, and remedial

measures; Hemisphere Pub. Corp.: Washington, 1985.

(67) Janz, G. J.; Allen, C. B.; Downey, J. R.; Tomkins, R. Physical properties data compilations

relevant to energy storage.I.Molten salts: Eutectic Data; Rensselaer Polytechnic Institute: New

York, 1978.